-
Notifications
You must be signed in to change notification settings - Fork 4.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug]: Agent doesn't read files bigger than 1.6k LINES of code. #5596
Comments
This is an interesting one. |
@LuciferMornens can you show exactly what the console output was when reading the large file? How do you know it's not reading the entire file? |
1.8k lines shouldn't have problems IMHO, it's not really large. 😅 I think the new Sonnet works fine with this size. It is limited to, I think 300-ish lines at a time, but that's okay, it just has to keep reading or modifying. @LuciferMornens can you please tell, what LLM and provider did you use? |
Hey. I can’t exactly provide you with the console output anymore as that was yesterday. @enyst it was Claude 3.5 sonnet new. And it clearly had an issue since it couldn’t see the last 200 lines of code in the file and I had to re prompt it a few times that it didn’t consider them and still didn’t do it. |
I'll keep an eye for that case! Please if you see it again, it will be great if we can see what commands and what errors it got. |
You betcha. I'll hit you with a reply here or a new issue if anything happens again regarding that. |
Is there an existing issue for the same bug?
Describe the bug and reproduction steps
Could possibly get a better way to read files with high amount lines of code.
So far i have a file with like 1.8k likes and it struggles with it, doesnt read it until the end as it should, even when it's trying to split it.
It is really a problem when working with bigger sized files.
Please 🙏 .
OpenHands Installation
Docker command in README
OpenHands Version
0.15.2
Operating System
WSL on Windows
Logs, Errors, Screenshots, and Additional Context
There weren't really issues, it just would never read the full file properly and it was failing getting that last content it needed.
The text was updated successfully, but these errors were encountered: